Missing Bit Depths?


Today’s Question: You addressed a question about bit depth and made reference to cameras with 12-bit and 14-bit support. But I thought images could only be either 8-bit or 16-bit?

Tim’s Quick Answer: While most software only offers options to work with images in either the 8-bit per channel or 16-bit per channel bit depths, there are a variety of other options available depending on the context.

More Detail: Bit depth refers to the level of precision involved in a conversion from an analog data source (such as light in the context of photography) to discrete digital values. In the context of digital photos, the low end of the bit depth scale is 8-bit per channel.

Most software provides support for either 8-bit per channel images or 16-bit per channel images. However, this is not an inherent limitation of image processing. If you capture a JPEG image with any digital camera, the result will be an 8-bit per channel image. With a raw capture, many photographers are familiar with the notion of a 16-bit per channel image.

However, just because most software supports 8-bit per channel and 16-bit per channel images doesn’t mean those are the only options for digital photos.

Again, for JPEG captures the image will always be an 8-bit per channel image. But for raw captures the camera is not necessarily producing a 16-bit per channel image. In fact, most are not. Most cameras offer either 12-bit per channel or 14-bit per channel analog to digital conversion. A few camera models offer full 16-bit per channel conversion.

However, regardless of the bit depth supported by your camera, in most cases when processing the raw capture to produce a final image, you will generally only have a 16-bit per channel option. In other words, whether your camera is producing 12-bit, 14-bit, or 16-bit per channel data, the result will be contained within a 16-bit per channel image file.

It is important to keep in mind that even if you are working with a 16-bit per channel image, that doesn’t mean you actually have 16-bit per channel data for that image. It is simply a matter of only having either an 8-bit per channel or 16-bit per channel option available in most software, so an image that has more than 8-bits per channel of data needs to be contained in a file that supports a higher bit depth, which generally translates to a 16-bit per channel file.

Note, by the way, that for HDR (high dynamic range) images, you might create a 32-bit per channel image. In other words, there are a variety of bit depth options depending on context, but in terms of digital images you will generally have 8-bit per channel and 16-bit per channel options available.